BTCRC-HN17-111, A phase 2 trial of ADT (goserelin) in combination with pembrolizumab for patients with advanced salivary gland tumors expressing androgen receptor (AR).

e18091 Background: Salivary gland cancers (SGC) are rare malignancies with no standard systemic therapy available for patients with recurrent or metastatic disease. Androgen receptors (AR) are expressed and functional on many SGCs. Preclinical data suggested possible synergy between ADT and immune checkpoint blockade. The Big Ten Cancer Research Consortium conducted an open label, multi-center, Simon’s two-stage, phase II trial using ADT in combination with pembrolizumab for advanced salivary gland tumors that express AR. Here, we report the preliminary results of the first stage of the trial. Methods: Patients with incurable recurrent or metastatic SGC (any histology) with at least 20% of tumor cells expressing AR by immunohistochemistry were recruited. Patients who had received prior ADT or immune checkpoint therapy were excluded. Patients were treated with subcutaneous injection of goserelin 3.6mg every 4 weeks. Pembrolizumab 200mg was administered intravenously every 3 weeks starting 14 days after initiation of goserelin. Disease assessment was performed after completion of 4 cycles of therapy. Primary endpoint was objective response rate. Key secondary endpoints included PFS, OS, and safety of the combination. Stopping rules for excess toxicity were employed. We planned enrollment of 9 patients in the initial stage, and plan to complete enrollment if we observed³ 2 objective responses. Results: Between June 2019 and June 2021, we enrolled 9 patients into the study to complete the first stage. 7/9 were men, 8/9 were white, and the median age was 67 years old. A total of 5 ductal carcinomas, 3 adenocarcinomas, and 1 adenoid cystic carcinoma patients were enrolled. Patients had a median of 1 prior line of treatment. No stopping boundaries were crossed for excess toxicity. One patient experienced Grade 4 Stevens-Johnsons syndrome requiring permanent discontinuation of pembrolizumab. Otherwise, there was no increase in immune-related toxicity from what might be expected with pembrolizumab alone. Objective responses were observed in 2/9 patients (22%). Both responses were seen in ductal carcinomas. The best response was partial response in 2/9, while 6/9 had stable disease as best response for clinical benefit rate of 88% (8/9). The 6-month PFS was 63% (CI 19-85%). Conclusions: The combination of goserelin and pembrolizumab was well-tolerated and show preliminary evidence of efficacy in the first stage of the trial. The second stage of the trial has opened to accrual with a plan to enroll 11 additional patients. Clinical trial information: NCT03942653.